The term “junk DNA” was coined by Susumu Ohno in conference presentations at the Brookhaven National Laboratory in Upton, New York and in Rhein, Germany, which were printed in conference proceedings volumes a short time later (Ohno 1972, 1973). Ohno used the term “junk DNA” only once per article (in the titles), and most of the argument focused on his view that gene duplication was necessary for evolutionary change, that a small fraction of duplicated genes would mutate in such a way as to acquire novel protein-coding functions and consequently that for every new gene that emerged a large number of non-functional gene copies would result, and that indeed much of the genome cannot be directly essential or any mutation anywhere in the genome would be deleterious and the organisms would not withstand even a mild mutation rate.
My suspicion is that most people who cite these papers have not read them, because they tend to do so by citing these (especially Ohno 1972) as “the” claim, supposedly adopted generally soon thereafter, that noncoding DNA was totally unimportant and should be dismissed in favour of studying protein-coding sequences. This history of the field is totally incorrect, as I have attempted to show by referring directly to the primary literature from the relevant period of the 1970s and ’80s. (Notably, creationists seem to think the discussion of the topic began in the 1990s, which is the only way they can argue that “intelligent design” predicted function when “Darwinists” rejected it).
I will be posting a series of excerpts from papers written in the 1970s soon. My point, as previously, is that up to and beyond Ohno’s proposed mechanism for generating “junk DNA”, the standard assumption, based on adaptationism (i.e., “Darwinism” under the proper definition), was that noncoding sequences were functional. The current view, that much or even most (but certainly not all) noncoding DNA is probably not adaptive for the organism, had to arise against the current of adaptationist assumptions and survived only because it is supported by far more evidence than the alternative.
For now I simply will give some indication of the literally immediate reaction to Ohno’s suggestion, recorded as a transcript of the discussion following his Rhein presentation and printed after the text of Ohno (1973). Here it is, beginning at the section relating to Ohno’s presentation:
EVANS: Dr. Ohno, I suppose the take-home message was that in your view certainly the highly repetitive DNA is junk and you said that lots of unique sequences are also junk. This point of view is now open to discussion.
YUNIS: I wonder if you really mean “junk”. You are equating non-translational and non-transcriptional DNA with junk. I agree that you must be right up to some extent, but I wonder whether you have ignored the proven polyploidization as a way of evolution.
OHNO: If there is any gene which is doing some good for your general well-being, you will suffer when you lose that gene. For this very reason a fraction of randomly sustained mutations of that locus would be deleterious. There is simply no way of having a useful gene without paying a certain price for the cost of natural selection. If, on the other hand, there is a gene which is totally irrelevant, you will lose that gene sooner or later, for natural selection would not police that gene.
YUNIS: We know that constitutive heterochromatin is rich in repetitive DNA and that satellite DNA spaces essential regions such as the centromere and nucleolar organizer. Isn’t this an important role?
OHNO: Yes, spacer is important in the same negative way as fribrinopeptides A and B. Only a short stretch of base sequence at its end would have to be conserved as a signal to be nicked by ribonuclease.
HENNIG: I feel one could accept to some extent both views. From all what is known so far we can conclude that probably the nucleotide sequence as such does not matter. Furthermore the actual amount of simple sequence DNA (within some limits) seems not to be important. But since this kind of DNA is there one has to correlate it to some kind of function. That means that either simply the presence of some portion of this material is essential for structural or functional reasons. Or one could imagine that this kind of DNA is a product of certain molecular mechanisms which as such are essential for the eukaryote genome. This, of course, should be some kind of multiplication mechanism. Such a mechanism may, for example, exist in order to keep ribosomal cistrons or 5S cistrons etc. by occasional multiplications alike. Such multiplication steps could, accidentally or not, include adjacent DNA sequences, thus producing simple sequence DNA. It is remarkable that nucleoli usually are associated with heterochromatin which contains simple sequence DNA. Also the histone genes in Drosophila seem to be associated with heterochromatin as Dr. Pardue has shown.
YUNIS: It is important to emphasize that, in general, there is a certain constancy in the amount and distribution of satellite DNA.
HENNIG: It is certainly true that there is a constancy in certain limits of simple sequence DNA. But these limits could be governed by simple mechanical requirements of the chromosomes, for example in segregation of the chromosomes. Extremely large chromosomes do have difficulties during cell division and thus an upper limit could be introduced by the size of chromosomes. I think the variability of the heterochromatic arm of the X chromosome of the Drosophila species, which we are studying, is a good example for a block of simple sequence DNA which seems not to be essential. Deletion of the long heterochromatic arm of the X in D. hydei has no obvious consequences for the flies. This could mean that there is some DNA which may not be necessary but is there and is kept. Of course, this deletion stock has not been tested for its success in a population competition with the wild type.
FORD: I think the word “junk” is a powerful word. The only thing I would seriously question is this assumption that perhaps just 10 units would be sufficient to act as a spacer when 100 or more were there. Do we really know enough to be sure of that point?
OHNO: If we argue that a given spacer can change its base sequence any way it likes, but that a length of it has to be conserved rigidly, deletions would become deleterious to spacer function. It follows that spacer sequences, too, contribute to the overall mutation load, and for this very reason, we cannot even afford to keep too many spacers.
FORD: I think it just wouldn’t be there unless it would do it. Something was a functional reason of some kind for it.
YUNIS: This is what I emphasized earlier, that this DNA must have a functional value since nothing is known so widespread and universal in nature that has proven useless.
FRACCARO: Well, there is an exception to that rule. A lot of us have permanent positions at the University but are considered by others (mainly by students) meaningless and of no utility whatsoever.
EVANS: Well with that very splendid comment I think we should now draw today’s discussion to a close. I should remind you, however, that we have at least a full hour available tomorrow to continue this discussion and would like to end by thanking all the speakers for their excellent presentations and the discussants for talking part in the discussion.
Part of the Quotes of interest series.
Ohno, S. 1972. So much “junk” DNA in our genome. In: Evolution of Genetic Systems (ed. H.H. Smith), pp. 366-370. Gordon and Breach, New York.
Ohno, S. 1973. Evolutional reason for having so much junk DNA. In: Modern Aspects of Cytogenetics: Constitutive Heterochromatin in Man (ed. R.A. Pfeiffer), pp. 169-173. F.K. Schattauer Verlag, Stuttgart, Germany.